Speeding up Tabular Reinforcement Learning Using State-Action Similarities.
Ariel RosenfeldMatthew E. TaylorSarit KrausPublished in: AAMAS (2017)
Keyphrases
- state action
- reinforcement learning
- evaluation function
- action space
- continuous state
- function approximation
- state space
- function approximators
- average reward
- temporal difference
- markov decision process
- similarity measure
- learning algorithm
- stochastic games
- markov decision processes
- optimal policy
- policy gradient
- reinforcement learning algorithms
- machine learning
- action selection
- neural network
- model free
- belief state
- learning automata
- state transitions
- multi agent
- transition probabilities
- optimal control
- policy iteration
- black box
- real valued
- supervised learning
- learning process
- search space